Analysis of Market-Ready Traffic Sign Recognition Systems in Cars: A Test Field Study

Advanced Driver Assistance System (ADAS) represents a collection of vehicle-based intelligent safety systems. One in particular, Traffic Sign Recognition System (TSRS), is designed to detect and interpret roadside information in the form of signage. Even though TSRS has been on the market for more than a decade now, the available ones differ in hardware and software solutions they use, as well as in quantity and typology of signs they recognize. The aim of this study is to determine whether differences between detection and readability accuracy of market-ready TSRS exist and to what extent, as well as how different levels of “graphical changes” on the signs affect their accuracy. For this purpose, signs (“speed limit” and “prohibition of overtaking”) were placed on a test field and 17 vehicles from 14 different car brands underwent testing. Overall, the results showed that sign detection and readability by TSRS differ between car brands and that even small changes in the design of signs can drastically affect TSRS accuracy. Even in a controlled environment where no sign has been altered, there has been a 5% margin of misread signs.

[1]  Yizhou Yu,et al.  Traffic Sign Detection Using a Multi-Scale Recurrent Attention Network , 2019, IEEE Transactions on Intelligent Transportation Systems.

[2]  Faliang Chang,et al.  Machine Vision Based Traffic Sign Detection Methods: Review, Analyses and Perspectives , 2019, IEEE Access.

[3]  Monika Sester,et al.  Traffic Regulator Detection and Identification from Crowdsourced Data - A Systematic Literature Review , 2019, ISPRS Int. J. Geo Inf..

[4]  Meng Lu,et al.  Technical Feasibility of Advanced Driver Assistance Systems (ADAS) for Road Traffic Safety , 2005 .

[5]  Feng Xiao,et al.  Improved Traffic Sign Detection and Recognition Algorithm for Intelligent Vehicles , 2019, Sensors.

[6]  Rung Ching Chen,et al.  Weight analysis for various prohibitory sign detection and recognition using deep learning , 2020, Multimedia Tools and Applications.

[7]  E. Hoffmann,et al.  Drivers' awareness of traffic sign information , 1991 .

[8]  Caroline Crump,et al.  Driver Distraction and Advanced Vehicle Assistive Systems (ADAS): Investigating Effects on Driver Behavior , 2017 .

[9]  Staniša Milošević,et al.  Presentation factors and driver characteristics affecting road-sign registration , 1986 .

[10]  Anjan Gudigar,et al.  A review on automatic detection and recognition of traffic sign , 2014, Multimedia Tools and Applications.

[11]  Omer Faruk Bay,et al.  Neural Network Based Vehicular Location Prediction Model for Cooperative Active Safety Systems , 2018 .

[12]  Kevin Heaslip,et al.  Prediction of traffic sign vandalism that obstructs critical messages to drivers , 2017 .

[13]  Hilario Gómez Moreno,et al.  Traffic sign shape classification and localization based on the normalized FFT of the signature of blobs and 2D homographies , 2008 .

[14]  Fernando Garcia,et al.  A Review of Sensor Technologies for Perception in Automated Driving , 2019, IEEE Intelligent Transportation Systems Magazine.

[15]  Aini Hussain,et al.  Vision-Based Traffic Sign Detection and Recognition Systems: Current Trends and Challenges , 2019, Sensors.